Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
Measuring the effect of distractions in a gaze-tracked
video-game
Helena Esborn Hejendorf Hansen
[email protected], Aalborg University, Denmark
Casper Høgalmen
[email protected], Aalborg University, Denmark
May 25, 2021
Abstract
Eye-tracking in video-games is proven to be a rapidlygrowing technology amongst developers and researchers,but is yet to reach the average video-game consumer.Largely used as an optional supportive element, eye-tracking affords game-play mechanics such as enemytagging and interaction at gaze. Video-games often of-fer a multitude of tasks to navigate, which can be dif-ficult to navigate utilizing only eye-tracking input. Inan attempt to explore the eye-tracking technology fur-ther, a gaze-controlled Breakout game was developed toreveal how different levels of distractions impact users’task-load, performance and game experience. The gameis evaluated utilizing within subject design and self-reported questionnaires which are collected in-betweeneach distraction spawn-rate. The results indicated thatthe user might be more aware of the key-objects of thegame once simultaneously presented with a large num-ber of distractions. It can therefore be concluded thatones focus will differ at a higher distraction rate, henceignoring the distractions to a degree. Further prelimi-nary study would have to be done in order to unveil thethreshold of when participants’ task-loads and game ex-periences change significantly.
1 Introduction
Eye-tracking technology has undergone rapid growth interms of both popularity and refinement amongst re-searchers and developers. Used for a variety of differ-ent disciplines, eye-tracking has among them been intro-duced into video-games [28]. It is often utilized as anoptional setting to support various features e.g. aim atgaze and enemy tagging. Eye-tracking in video-gamescan be described as a supportive element to the mainintention of the system as opposed to a main featurein and of itself. Examples of commercial video gamesthat utilize eye tracking as an optional feature includeAssassins Creed Valhalla, Shadow of the Tomb Raider,
Far Cry 5 and The Division 2 [1].
In terms of research, many findings exist on the topicof eye-tracking hardware and its utilization. Investi-gations into eye-tracking as a game mechanic, and theeffect of utilizing it as such, has been explored in dif-ferent studies, either by modifying an existing commer-cial game or by creating entirely new games with eye-tracking as a main mechanic. However, its utilization incommercial video-games can still be addressed as beingunder-explored [9].
Research suggests, that eye-tracking, and gaze-basedinteractions in general, have a compelling future invideo-game. A survey by Velloso et. al. investigatedthe future of eye-interaction in video-games, which high-lights the possibilities for gaze-enabled games. Theyfocus on the relevance of continuing to study and in-vestigate the topic of eye-tracking and gaze-enabledgames, especially with eye-tracking as the main gamemechanic. This includes studies on the limitations anddisadvantages which follow, compared to traditional in-put modalities such as mouse and keyboard, or a con-troller [25].
Gaze-controlled games have also been proven toyield a heightened sense of immersion and flow for theplayers. Several studies have investigated and utilizedbetween-subjects design for their evaluation. Herethe eye-tracking input has consistently yielded morepositive results compared to traditional mouse andkeyboard controls, in terms of immersion, presence andflow [23, 24, 15].
However, games are also known to present a plethoraof simultaneous tasks to players. Whether tacticalor strategical, players are invariably presented withmultiple mental processes at once and expected tomanage them in order to progress. Taking traditionalfirst-person shooting combat as an example, playersare required to keep track of their own health, aimtheir gun at moving enemies, tracking enemies in thesurrounding environment and move their character
1
accordingly etc. In essence, players are required tomultitask - a process which according to Huesteggetakes its toll on performance costs [7].
Connecting multitasking to increased response timesand error rates, Huestegge also found that saccades cantrigger the multitasking phenomena, treated as a task initself by the human mind for all intents and purposes.Faced with multiple tasks in multiple places, partici-pants were slower to initiate a saccade or manual re-sponse when they were planning another movement atthe same time [7].
This highlights an issue in video-games where eye-tracking is used as a main mechanic; if several distrac-tions that might require attention are presented, theplayer must utilize multitasking to stay focused on thegoal of the game, which may lead to a performance cost.
This paper aims to contribute with a further look intowhat limitations and disadvantages there might be withgaze-enabled games. We investigated the influence ofdistractions in gaze-controlled games, for which we de-veloped a gaze controlled Breakout game with the goalof distracting the player with several positive and neg-ative power-ups. The effects on the players gaze andperformance will be measured through questionnairesand through observing their scores, amount of loses andobjects logging.
2 Related Work
Multitasking is described as the performance of multi-ple tasks concurrently [13] [10]. This definition origi-nates from computer terminology but has been appliedto human behavior for multiple tasks at once, such asdriving and talking on the phone or attending a classand taking notes [10, 13].
It has been proposed by Pashler, that during the pro-cess of multitasking, a mental process will occur in onetask. While this process transpires, a similar processcannot occur in any other task simultaneously; a bot-tleneck effect [3][17].
Eye-movements can cause multitasking issues, for ex-ample Huestegge et al. states that saccades both exhibitand cause dual-task costs in the context of other actionsand should thus also be regarded as a response modality- A conveyor of information from the body as well as agatherer of visual information from the world [7]. Thisdual-task cost should be observed and kept in mind dur-ing our own evaluation, as this might result in a lowerperformance by the players.
Saccade is a term used to describe a type of eye move-ment, some of the most common ones being the follow-ing [21]:
• Saccades: Describes rapid eye movements used inre-positioning the fovea as the eyes jump from onefixation to the next.
• Fixations: Describes stabilizing the retina over astationary object, keeping it directly in view.
• Smooth pursuit: Describes keeping the eye fixatedon a moving target.
Velloso et. al. compiled and classified game mechan-ics that involve eyes from three different perspectives,which contribute with a practical toolbox which gamedesigners can utilize for their games [25]. They de-scribe game design solutions to (human computer in-teraction (HCI) problems, where they state that eventhough game interaction design can be seen as a subsetof HCI, designing interaction techniques for video-gamesfollows a different set of rules and goals. This is becausean interaction technique can’t necessarily be evaluatedby measuring completion time and error rates. Thisis due to these metrics not necessarily capturing socialfactors, the users’ state and other sensory, cognitive andbehavioural factors which may affect the game experi-ence [25]. They also state that challenging the playersnatural eye behaviour can create fun game mechanics,such as forcing players to utilize their peripheral vision.
However, it is important to recognize that a naturalmapping does not necessarily result in a positive gameexperience. Utilizing a paper by Dorr et. al. as an ex-ample, as they created a gaze-tracked Breakout game fora comparative study, where the player was controllingthe paddle to bounce a ball and break bricks [4]. Herethey found that by mapping the paddles x-coordinatesto the player’s gaze x-coordinates, all the players had todo was follow the ball with their gaze in order to progressin the game, which removed all challenge [4, 25].
At a trade fair show, Dorr et. al. presented theirgame to a visitor who had to experience with computergames. After performing very well, two minutes into thegame, she asked when the game was starting as she wasjust following the ball with her gaze, hence not noticingthe paddle following her gaze [4].
However, Dorr et. al. did find eye-tracking to be thesuperior input modality. Comparing the gaze-trackedversion of their Breakout game to mouse-controlledgame-play, the study revealed that the participantswho played the gaze-controlled version not only yieldedhigher scores, but also reported higher enjoyment of thecontrols. They conclude, that Breakout has a particu-larly simple and intuitive game play that makes it ide-ally suited for gaze playing. [4].
We will base our implementation on the findings ofDorr et. al, as they proved it was a game that wasable to get positive feedback from their evaluation uti-lizing eye-tracking [4]. However, 1:1 mapping with notbe implemented, but instead a delay on the paddlewill be utilized in order to expectantly avoid the issuesthat occurred in their implementation. Furthermore, aBreakout game will provide a solid foundation wherethe power-up distractions can easily be altered, as well
2
as the spawn-rate thereof and level layout. It lays thegroundwork for good and simple premise, which we caneasily be adjusted and wont need much explanation, inregards to game-play, to the participants.
Furthermore, Velloso et al. examined how the gamesinterprets eye-movements, which they describe as inputtypes, and they provide an overview of different gamemechanics which makes use of these modalities. Due toour implementation being based on real-time eye-basedinteraction, we will be using the continuous-only inputfrom their categorization, in which the X and Y coordi-nates of the users’ gaze-point are tracked continuously[25].
Gomez et al., argue for 5 different themes coveringtheir reflections of eye-tracking, provided by the resultsof their games, and related to other eye-tracking enabledgames. The themes consisted of attention to interaction,attention dilemmas, anti-intuitive gaze interactions, therole of metaphors and gaze identity and control. In thefirst theme, attention to interaction, Gomez et al., arguethat in gaze interaction one should avoid using tensionor only present one major interaction at a time. Theydo provide a counterpoint, in which the mobile game,Flappy Bird, one might also quickly lose, but they areencouraged to try harder in another attempt, in whichthey assume the same to be true if the game was gaze-controlled, stating that sometimes what is not intuitiveand natural can be fun and engaging [22]. In a gameof Breakout, it can be argued that the only major in-teraction the player has, is moving the paddle. How-ever, having to navigate different distractions in termsof power-ups which the player can utilize in order tostrategize can also be argued to be an interaction initself, hence leaning towards to example of Flappy Bird.
Gomez et al. also investigated and described differenttypes of gaze aversion of how this is utilized in video-games. They proposed a spectrum of five discrete cate-gorizes for unexpected use of the eye-tracking hardware,which are “might not look, “cannot look”, “should notlook”, “must not look” and “does not look”. The aim oftheir study was to unveil under-explored aspects of HCIthrough unexpected actions, and argues for a higher pe-ripheral awareness through such uses [22]. Gaze aver-sion also draws parallels to the concept of distracting theplayer utilizing game elements, which the player shouldattempt to avoid in order to focus on the task they havebeen given. Furthermore, this theory can be utilizedin context of distractions in a gaze-tracked version ofBreakout, as the player has to focus the majority oftheir attention on the paddle in order to not lose thegame. Hence, the various distractions can be argued tofall in the category ”should not look”.
Gaze controlled games has also shown to increase theplayers feeling of immersion. Vidal et al. investigatedimmersion in a gaze-tracked game by developing a video-game where the characters were designed to be reactive
to the players gaze, which yielded an increase in im-mersion due to the interactive gaze-controls [26]. Beeet al. also made a similar implementation, investigat-ing gaze behaviours in an interactive story-telling ap-plication. They tested the application with two groups,one of which tried the interactive version of the applica-tion and one which tried the non-interactive applicationwhere the characters would not react to the players gaze.They found that the interactive model afforded higherlevels of immersion [2]. Other researchers have gottensimilar results in their studies, where both immersion,engagement, entertainment, and sense of presence arepositively affected by gaze-based interactions in theirapplications [18, 5, 8, 15, 16, 27, 19, 11, 14].However, this is not the focus of our investigation, aswe are investigating the effect of distractions at differ-ent spawn-rates.
3 Design and Implementation
A game called ”Eyekanoid” was developed utilizing aTobii Eye Tracker and the Unity game engine. Basedon the critically acclaimed ”Arkanoid”, Eyekanoid hasthe player controls a paddle horizontally utilizing theireyes, bouncing a ball towards bricks in order to destroythem. Whenever a player destroys a brick, they areawarded 100 points and whenever the ball falls off thescreen, 500 points are detracted from their score. Thesescore values were determined through internal testing.The speed of the ball will gradually increase through-out the level, making each level harder and harder as itprogresses as players must evaluate the game state morerapidly with every speed increment.
The paddles x-coordinates is mapped to the x-coordinate of the players’ gaze, however, a delay, whichis relative to the movement, has been implemented inorder to avoid the issue encountered by Dorr. et. al.where the participants were simply following the ballwith their eyes in order to win the game [4].
For test-purposes, we define four tasks which maytake up cognitive capacity within a game of Eyekanoid:
• Paddle positioning: Positioning the paddle tocatch the ball as it falls.
• Target identification: Identifying where the ballis intended to go next. Tracked by looking at howlong players looked at the bricks.
• Ball tracking: Tracking the position of the balland its trajectory
• Power-up tracking: Identifying what power-upsare and evaluating whether they are worth goingfor.
3
As play progresses, power-ups will periodically fallslowly from the top of the screen. Catching these power-ups with the paddle will modify the game-state eitherpositively or negatively, depending on the power-upcaught. The spawn-rate of these power-ups is deter-mined by setting a number X as the spawn-rate. Oncethe game starts, the game will pick a float Y at randombetween 0 and X and subsequently start a timer. Whenthe timer reaches or exceeds Y, a random power-up fromamong the six available is spawned, weighted slightly to-wards the ‘Slow ball’ power-up. Once this occurs, thetimer is reset and a new Y is picked. The random na-ture of the power-ups is particularly worth noting asthe game experience each participant had, while simi-lar to that of their peers, was not identical. This wasdeemed a necessity to ensure that the power-up trackingtask was always performed as a reaction instead of as alearned response. For the power-up design there are 6different types of power-ups, each with their own prosand cons which players had to weigh in accordance withthe current game-state:
Figure 1: Top-left to bottom right: Slow ball, Fast ball,Small paddle, Large paddle , Destructor, Sticky Paddle.
• Slow Ball / Fast Ball: Decreases/Increases ballspeed. Makes the ball easier/harder to catchbut slows down/speeds up the process of breakingbricks.
• Small Paddle / Large Paddle: De-creases/Increases paddle width. Makes theball harder/easier to catch but decreases/increaseslikelihood of picking up unwanted power-ups.
• Destructor: Makes the ball to pass through brickswhen destroying them instead of bouncing off. Thiseffect lasts for 9 seconds. Allows for the destructionof multiple bricks in quick succession. Almost ex-clusively beneficial.
• Sticky Paddle: Makes the ball stick to the paddlefor two seconds before firing. This effect lasts for12 seconds. Allows a much larger degree of control
over ball positioning but slows down brick destruc-tion significantly.
The time which the effects lasts was determinedthrough internal testing.
The exact design of this effect line-up had multiplepurposes. Firstly, power-ups were intended to be anintegral game-play mechanic which players would peri-odically require to attain a good performance. The op-tion to ignore the power-ups completely would eliminatethem as ”distractions” for the purposes of our experi-ment. To ensure evaluation of the power-ups, the grad-ual speed increase of the ball was set high enough to ne-cessitate periodic pick-ups of the ”Slow ball” power-upin order for the participants to continue playing, with-out the speed of the ball being too fast. To furtherensure that power-ups were evaluated, some power-upswere made almost strictly negative. The ”Fast ball”power-up would speed up the ball-breaking process inthe short run, but, as the ball grew faster and playerspicked up more ”Faster balls” over the 80 second play-ing period, would end up being a detriment as playerswould in some cases not be able to catch the ball at all.Similarly, the ‘Smaller paddle’ power-up would decreasethe chance of catching the ball and catching the all-important ‘Slow ball’ power-up, making it an ill-advisedpower-up to catch. These negative power-ups furthernecessitated constant evaluation, as players could notsimply catch all power-ups caught in their peripherals,at the peril of catching some of the more negative ones.To further ensure active evaluation power-ups were allvisually represented as green circles with only differingblack icons on top. These icons depicting metaphors forthe effects of the power-up.
Based on this, three builds with different distractionspawn-rates were created:
• Build 1 (Slow): One Power-up every 5 seconds
• Build 2 (Medium): One Power-up every 3.5 seconds
• Build 3 (Fast): One Power-up every 2 seconds.
In addition, a system was implemented in each build totrack the gaze activity of the participant. This system isimplemented by each frame measuring the vector goingfrom the player’s gaze point to the closest point on eachgame object present in the level. While this system isroughly accurate, it depends on Tobii Eye-tracker gaze-point data, which is an average of an aggregation ofpossible points where the eye-tracker has determinedthe player may be looking. While accurate, this mayhave had an influence on the eye-tracking data gatheredduring the evaluation.
4 Evaluation
The evaluation was done using within subjects design,where all subjects are exposed to all conditions.
4
Figure 2: Eyekanoid
In our case, all participants were playing the sameversion of Eyekanoid utilizing the Tobii Eye Tracker 5,but were exposed to the changing amount of distractionsand level layouts in a different order. The order of thedistraction spawn-rates was created using Latin-squaredesign. This was done to balance out the order in whichthe participants were exposed to the distractions and toavoid a potential learning curve. The order looked likethe following:
Figure 3: Table of testing segments
Utilizing this testing method, a minimum of 12 partic-ipants were needed to have data from two participantson each order.
4.1 Evaluation method
The implementation was evaluated through both dataand observations from the participants playing thegame. These observations will consist of observing ifthe participants are seeking or avoiding the power-ups,eventual signs of frustration or enjoyment of the system,as well as verbal feedback during the duration of themplaying the game.
Our data was based on a demographic questionnaire,the NASA task load index (NASA-TLX), the in-gameiGEQ questionnaire by Kort et. al. and a system usabil-ity test [12, 6, 20]. All questionnaires were self-reportedby the user.
The NASA-TLX questionnaire is a subjective multi-dimensional assessment tool, which rates the perceived
workload of a task, system, performance or the effec-tiveness of a team [6]. This questionnaire was utilizedto measure if the task-load is reported to be differentwhen the participants are presented with a higher dis-traction spawn-rate.
The IGEQ questionnaire was be utilized to evalu-ate the participants game-play experience. The iGEQquestionnaire is a shorter in-game version of the GEQquestionnaire, which was developed in order to measureplayer experience multiple times during a game session[20].
Lastly, the system usability test was used to evaluatethe system itself. It can be an addition to evaluate theoverall experience of playing the game. For example, ifa user has low performance and reports low enjoymentof the system, the system usability scale might be usefulto determine if it is a fault of the system.
The total time participants looked at various objectsin the scene was also tracked and logged. The objectstracked are the following:
• The ball
• The Paddle
• The bricks
• Each type of power-up
Their score for each segment of the game was loggedas well.
The game was played in 80 second segments, with thegiven order of the distraction rates, with the NASA-TLX and iGEQ questionnaire reported in-between eachsegment. The structure looked like the following:
• Participant plays the first segment of 3 levels withone distraction spawn-rate.
• Participant answers the NASA-TLX and iGEQquestionnaires.
• Participant plays the second segment of 3 levelswith one distraction spawn-rate.
• Participant answers the NASA-TLX and iGEQquestionnaires.
• Participant plays the third segment of 3 levels withone distraction spawn-rate.
• Participant answers the NASA-TLX and iGEQquestionnaires, as well as the SUS-test.
This structure was utilized to ensure we got dataon how each participant experienced the spawn-ratefor the power-ups and how they were affected by themas distractions. On the basis of this, we created thefollowing hypotheses:
5
Hypothesis 1 For hypothesis 1 we theorize thatthere will be a difference in players’ performances whenfaced with different levels of distractions. This will bemeasured by looking at player scores for each of thethree game versions. Based on Huestegge’s researchon distractions and their influence on gameplay, it isexpected that as distraction levels rise, player scoredecreases.
H01 : µhighspawnrate = µmediumSpawnrate = µlowSpawnrate
H11 : µhighspawnrate 6= µmediumSpawnrate 6= µlowSpawnrate
Hypothesis 2 For hypothesis 2 we theorize that therewill be a difference in the perceived physical and mentaltask load when facing players with different levels ofdistraction. This will be investigated through the NASAtask load index questionnaire. Based on the research of(?) we expect the mental task load to increase as thelevel of distractions increase.[6].
H02 : µhighspawnrate = µmediumSpawnrate = µlowSpawnrate
H12 : µhighspawnrate 6= µmediumSpawnrate 6= µlowSpawnrate
Hypothesis 3 For hypothesis 3 it is theorized thatthere is a significant difference between player’s game-play experiences when faced with different levels of dis-traction. This will be measured through the in-gameGameplay Experience Questionnaire (iGEQ) which in-quires about a player’s experience playing the game. Itis here expected that a higher level of distractions willyield a more frustrating or engaging experience due toa difference in difficulty. This is expected as a result ofthe player’s flow fluctuating. [20].
H03 : µhighspawnrate = µmediumSpawnrate = µlowSpawnrate
H13 : µhighspawnrate 6= µmediumSpawnrate 6= µlowSpawnrate
4.2 Participants
Our evaluation group was young males and femaleswho are familiar with video-games, as this target groupwill most likely already be familiar with the concept ofBreakout as well as having experience with multitask-ing in video-games. We had a total of 12 participantsbetween 20-27 years of age. They consisted of 10 malesand 2 females, all playing video-games 25,75 hours perweek on average. All participants were reported to befamiliar with the concept of eye-tracking but to havelittle firsthand experience with it. Three participantsreported to have utilized it before to varying degrees.
4.3 Procedure
At the beginning of each testing session, the participantwas asked to fill out a consent form stating that thedata will only be saved for the purpose of this project.
Once signed, they were asked to fill out a demographicsquestionnaire, in which they would state their age, howmany hours they play video-games per week, as well astheir familiarity and experience with eye-tracked games.Once finished, the participants would get an introduc-tion to what the game is about and how to play it. Theywould also be told that the speed of the ball would in-crease linearly and that they could use the ”Slow Ball”power-up to manage the speed of the ball. The onlypower-ups not explained were the ”Sticky Paddle” and”Destructor” power-ups to keep a sense of exploration inthe game. The participant was also encouraged to thinkaloud whilst playing. Afterwards they would be set toplay the game, answer the self-reported questionnairesafter each segments and at the end of the evaluationanswer a SUS-test.
4.4 Results
The data gathered from the evaluation was averaged foreach participant across all segments for more efficientanalysis. These averages were then utilized to performa statistical analysis utilizing a two-factor ANOVA-testwith the distraction level acting as the independent vari-able.
Figure 4: Data of object logging and scores
The low P-value (P ¡ 0,05), which can be observedfor the ball, paddle, the large paddle power-up, meansthat there is not enough evidence to prove that the dis-traction spawn-rates are not equal. For the remain-ing logged gaze-points, a significance is proven, whichsuggests that there is a significance difference in whatthe participants were looking at, at the different spawn-rates.
The averages across all of the participants’ object log-ging and scores showed that the participants in generallooked at the objects more, specifically the paddle andthe ball, and also yielded the highest scores at the fastestdistraction spawn-rate.
6
Figure 5: Average result for the low spawn-rate
Figure 6: Average results for the medium spawn-rate
Figure 7: Average results for the fast spawn-rate
The two-factor ANOVA-test with the distraction levelacting as the independent variable, was also performedon the iGEQ-questionnaire and the NASA-TLX. Butsignificant difference was found in either of the ques-tionnaires.
For both of the questionnaires, we have p-valueshigher than our significance value (0,05), which meansthat means that we don’t have enough evidence to provethat the distraction spawn-rate means are not equal.
The SUS-test yielded a score of 61,75, which can beinterpreted as an average system usability score.
5 Discussion
5.0.1 Hypothesis 1: Their scores
This hypothesis aimed at investigating the difference inthe participants score, and assumed that the partici-pants will yield a lower score in the segment with thehigh spawn-rate of distraction. For the slow distrac-tion spawn-rate segments the mean of the participantsscores was 3058,33 points with a standard deviation of3424,776. For the medium spawn-rate segments the par-ticipants got a mean of 2383,33 points with a standarddeviation of 5644,441. The high standard deviation in-dicates that the values are spread out on a higher rangeand that the data may not be normally distributed.Lastly, for the fast distraction spawn rate the partici-
pants got a mean of 3925 points with a standard devia-tion of 3362,321.
This can be due to the participants being more fo-cused on the objects around them in the game, as theresults indicate that the participants spent a lot moretime gazing at the ball and the paddle. Once too manydistractions were present on the screen at once, the par-ticipants seemed to change their strategy to focus on thegame-elements required to succeed. Participants acquir-ing a higher score through this method could indicatea flaw in the game design as the ”slow” power-up wasintended to be essential. For the slow distraction spawn-rate they averagely gazed at the ball and the paddle for13,73 seconds and 29,13 seconds respectively. In themedium spawn-rate segment they gazed at the ball foraveragely 14,91 seconds and the paddle for 37,32 sec-onds. However, in the fast distraction spawn-rate seg-ment they gazed at the ball and the paddle for 29,10seconds and 67,77 seconds respectively, which is a sig-nificant increase. While difficult to decisively conclude,it is believed that when faced with a sufficient numberof power-ups, players prioritized elements most essentialto a good performance; the ball and the paddle. Theoccurrence of logging numbers similar to other distrac-tion levels is thus theorized to be a result of increasedpower-up frequency corresponding to more power-upsentering players’ focuses by chance.
5.0.2 Hypothesis 2: NASA-TLX questionnaire
Hypothesis 2 assumed that it will be more physicallyand mentally demanding for the participants to playthe game with a higher spawn-rate of distractions, whichwas investigated through the NASA-TLX questionnaire.The NASA-TLX questionnaire proved that there was nosignificance between the different spawn-rates, meaningthat the participants reported no significant differencein the task-load. This is could be due to an issue inour methodology, as the difference in the spawn-rateswere not big enough to impact their task-load signifi-cantly. This could be avoided by doing more in-depthpreliminary testing with more participants, testing thethreshold of where ones’ task-load changes more exten-sively. However, adding more power-ups may not makea significant change to what objects the participants arelooking at or for how long, but could have had an impacton their performance by giving them access to more in-stances of the positive power-ups such as ”Destructor”.
5.0.3 Hypothesis 3: iGEQ in-game question-naire
This hypothesis assumed that the participants wouldfeel more challenged playing the game with the higherdistraction spawn-rate, which was unveiled through theiGEQ in-game questionnaire. The iGEQ in-game ques-tionnaire yielded no significant difference between the
7
different spawn-rates, which means that the participantsfeelings doesn’t change significantly throughout the dif-ferent segments. This could also have been avoidedthrough more extensive preliminary testing, learning atwhat spawn-rate the participants feels the most skillfuland challenged. This could also have helped us establisha better flow in the game overall, as well as pointing toat which spawn-rate the most sense of flow occurs forthe participants.
5.0.4 Other findings
During the evaluation, it was observed, that despitebeing encouraged to think aloud during the testingprocess, most participants were quiet and focused onthe game. This could have been due to the game beingchallenging to the participants, but according to theanalysis done on the iGEQ in-game questionnaire,there was no significance regarding their feelings orchallenge towards the different spawn-rates. Only fivetimes out of the total 36 segment reports from all theparticipants, was it reported that they felt ”extremely”challenged.
It was also observed, that the participants weremostly seeking the ”slow” and ”Destructor” power-upand avoiding the ”fast” and ”sticky” power-up. Seekingthe ”slow” power-ups, which decreases the speed of theball, is a good strategy against the linear increase in thespeed of the ball. Collecting the ”fast” power-up couldlead to a perpetual inability to catch the ball later inthe game, which makes sense to why it was avoided.The ”Destructor” power-up was also particularly usefulfor the participants, as it allowed them to gain morepoints, as the ball would not bounce off the bricks. Thispower-up might also have been a determining factoras to why the participants yielded a higher score forthe fast distraction spawn-rate. All of the power-upswould spawn more often, leading to the participantsalso being able to pick up this power-up more often.
The insignificant data could be caused by issuesin our methodology. Having participants being apart of our preliminary testing could have helped usdetermine the values of the distractions rates withmore participants, as we would be able to measure atwhat distraction spawn-rates one’s task-load increasesor decreases. These values could then have beenused in the evaluation, which most likely would havegiven us more significant results, whilst providing uswith the data needed to conclude when an amount ofdistractions has a significant impact on ones task-loadand game experience.
Overall, the data that was gathered from the ques-tionnaires was very subjective to each participants expe-rience, and could therefore not describe when an amount
of distractions affects ones task-load or enjoyment of thegame.
6 Conclusion
The aim of this project was to investigate the impact ofseveral levels of distractions in a gaze-controlled Break-out game. The distractions was presented as both pos-itive and negative power-ups, which the player wouldhave to navigate through and strategize which to pickup to their benefit of yielding a higher score. The partic-ipants score, how long they looked at the game objectswas logged, and the participants task-load as well astheir feelings towards the game experience was reportedthrough the NASA Task Load Index and the iGEQ in-game questionnaire respectively. The results showed,that whilst no statistical significance was found througheither of the questionnaires, the object logging did showa significantly higher value in terms of how long the par-ticipants looked at the paddle and the ball when playingat the highest distraction spawn-rate. Therefore, it maybe concluded that the participants were more focused onthose two key game-objects that are crucial in order tosucceed, but that the spawn-rates did not differ enoughfrom each other to demonstrate when the amount of dis-tractions had a significant impact on the participants’perceived task-load and game experiences.
7 COVID-19 disclaimer
Due to the current state of society during the COVID-19 pandemic when this projects was conducted, the ex-periments had to be changed accordingly. This yieldeda number of of negative-effects on the process of designand implementation of the product. The campus of Aal-borg University was shut down for the majority of theproject and student were encouraged to work separatelyfrom home. This had an impact on our group coordina-tion and collaboration, as all communication had to beconducted online. An example of this, is that only onemember of the group had access to the one eye-trackerwe were given for the project, making the implemen-tation especially difficult. This also meant that moreor less all the tasks and work of this project had tobe divided between the members, thus minimizing theamount of collaboration between the members of thegroup. This resulted in a lot of planning and internaltesting, to make sure that everyone’s expectations andvisions for the final product was reached. This was donethrough screen-sharing and giving the other member fre-quent updates on their work progress, as well as adjust-ing game-play values along the way. It also affected theevaluation of the project, as one member is due to pre-existing conditions, a high-risk for COVID-19, meaningthat the evaluation was mostly carried out by the other
8
member, at university campus, when it was opened for ashort period of time. It also resulted in lower amount ofparticipants than what would be desired under normalconditions as well as limiting our preliminary testing,thus making it difficult to draw any meaningful conclu-sions. This method of working, in general led to a lot ofpre-planning and meetings, doing our best to keep eachother updated, so that the quality of the project wasmaintained. .
References
[1] Tobii Gaming — PC Games with Eye Tracking.Top Games from Steam, Uplay.
[2] Bee, N., Wagner, J., Andre, E., Vogt, T.,Charles, F., Pizzi, D., and Cavazza, M. Dis-covering eye gaze behavior during human-agentconversation in an interactive storytelling applica-tion. In International Conference on MultimodalInterfaces and the Workshop on Machine Learn-ing for Multimodal Interaction, ICMI-MLMI 2010(2010).
[3] Borst, J. P., Taatgen, N. A., and van Rijn,H. The Problem State: A Cognitive Bottleneck inMultitasking. Journal of Experimental Psychology:Learning Memory and Cognition 36, 2 (mar 2010),363–382.
[4] Dorr, M., Pomarjanschi, L., and Barth,E. Gaze beats mouse: A case study on a gaze-controlled breakout. Tech. Rep. 2, 2009.
[5] Gowases, T., Gowases, T., Bednarik, R.,and Tukiainen, M. Gaze vs. Mouse in Games:The Effects on User Experience.
[6] Hart, S. G. NASA-task load index (NASA-TLX);20 years later. In Proceedings of the Human Factorsand Ergonomics Society (nov 2006), SAGE Publi-cationsSage CA: Los Angeles, CA, pp. 904–908.
[7] Huestegge, L. The role of saccades in multitask-ing: Towards an output-related view of eye move-ments. Psychological Research 75, 6 (nov 2011),452–465.
[8] Hulsmann, F., Dankert, T., and Pfeiffer,T. Comparing Gaze-based and Manual Interactionin a Fast-paced Gaming Task in Virtual Reality.Tech. rep.
[9] Isokoski, P., Joos, M., Spakov, O., and Mar-tin, B. Gaze controlled games. In Universal Accessin the Information Society (2009), vol. 8, pp. 323–337.
[10] Kenyon, S., and Wing, G. A. What do we meanby multitasking?-Exploring the need for method-ological clarification in time use research. Tech.Rep. 1, 2010.
[11] Lankes, M., Mirlacher, T., Wagner, S., andHochleitner, W. Whom are you looking for?the effects of different player representation rela-tions on the presence in gaze-based games. In CHIPLAY 2014 - Proceedings of the 2014 Annual Sym-posium on Computer-Human Interaction in Play(oct 2014), Association for Computing Machinery,Inc, pp. 171–179.
[12] Lewis, J. R. The System Usability Scale: Past,Present, and Future. International Journal ofHuman-Computer Interaction 34, 7 (jul 2018), 577–590.
[13] Madore, K. P., and Wagner, A. D. Multicostsof Multitasking. Cerebrum : the Dana forum onbrain science 2019 (2019).
[14] Nacke, L. E., Kalyn, M., Lough, C., andMandryk, R. L. Biofeedback game design: Usingdirect and indirect physiological control to enhancegame interaction. In Conference on Human Fac-tors in Computing Systems - Proceedings (2011),pp. 103–112.
[15] Nacke, L. E., Stellmach, S., Sasse, D., andLindley, C. A. Gameplay experience in a gazeinteraction game.
[16] Nielsen, A. M., Petersen, A. L., andHansen, J. P. Gaming with gaze and losing witha smile. In Eye Tracking Research and ApplicationsSymposium (ETRA) (New York, New York, USA,2012), ACM Press, pp. 365–368.
[17] Pashler, H. Processing stages in overlappingtasks: Evidence for a central bottleneck. Journal ofExperimental Psychology: Human Perception andPerformance 10, 3 (1984), 358–377.
[18] Pashler, H., and Hoffman, J. Saccadic EyeMovements and Dual-task Interference. The Quar-terly Journal of Experimental Psychology SectionA 46, 1 (feb 1993), 51–82.
[19] Perreira da Silva, M., Courboulay, V., Pri-gent, A., Prigent Gameplay, A., PerreiraDa Silva, M., Courboulay, V., and Prigent,A. Gameplay experience based on a gaze trackingsystem The 3rd Conference on Communication byGaze Interaction-COGAIN 2007: Gaze-based Cre-ativity Gameplay experience based on a gaze track-ing system. Tech. rep., 2007.
9
[20] Poels, K., De Kort, Y. A. W., and Ijs-selsteijn, W. A. 028765 FUGA The fun ofgaming: Measuring the human experience of me-dia enjoyment STREP / NEST-PATH DeliverableD3.3: GAME EXPERIENCE QUESTIONNAIRETitle of Contract FUGA-The fun of gaming: Mea-suring the human experience of media enjoymentAcronym FUGA Con. Tech. rep., 2006.
[21] Purves D, Augustine GJ, F. D. Neuroscience.2nd edition: Types of Eye Movements and TheirFunctions. Sinauer Associates, Sunderland (MA),2001.
[22] Ramirez Gomez, A., and Gellersen, H. Look-ing Outside the Box: Reflecting on Gaze Inter-action in Gameplay. In Proceedings of the An-nual Symposium on Computer-Human Interactionin Play (New York, NY, USA, oct 2019), ACM,pp. 625–637.
[23] Smith, J. D., and Graham, T. C. Use of eyemovements for video game control. In InternationalConference on Advances in Computer Entertain-ment Technology 2006 (New York, New York, USA,2006), ACM Press, p. 20.
[24] Spakov, O., Istance, H., Raiha, K. J., Viita-nen, T., and Siirtola, H. Eye gaze and headgaze in collaborative games. In Eye Tracking Re-search and Applications Symposium (ETRA) (NewYork, NY, USA, jun 2019), Association for Com-puting Machinery, pp. 1–9.
[25] Velloso, E., and Carter, M. The emergenceof eyeplay: A survey of eye interaction in games.In CHI PLAY 2016 - Proceedings of the 2016 An-nual Symposium on Computer-Human Interactionin Play (oct 2016), Association for Computing Ma-chinery, Inc, pp. 171–185.
[26] Vidal, M., Bismuth, R., Bulling, A., andGellersen, H. The royal corgi: Exploring socialgaze interaction for immersive gameplay. In Con-ference on Human Factors in Computing Systems- Proceedings (apr 2015), vol. 2015-April, Associa-tion for Computing Machinery, pp. 115–124.
[27] Wilcox, T., Evans, M., Pearce, C., Pollard,N., and Sundstedt, V. Gaze and voice basedgame interaction: The revenge of the killer pen-guins. In SIGGRAPH’08 - ACM SIGGRAPH 2008Posters (New York, New York, USA, 2008), ACMPress, p. 1.
[28] Yousefi, M. V., Karan, E. P., Mohammad-pour, A., and Asadi, S. Implementing EyeTracking Technology in the Construction Process.Tech. rep., 2015.
10
A Appendix A
A.1 Participant Results and Data
11
Participants – Raw logged data:
Participants raw logged data for the number of seconds they were looking at game elements
All units are in seconds.
Participant 1 – slow:
Level 1 Gaze Logging
Ball: 8.393653
Racket: 25.28928
Blocks: 13.34774
Slow: 0.6831629
Fast: 1.015843
Enlarge: 1.61639
Shrink: 0
Destructor: 0
Sticky: 0
Level 1 Gaze Logging
Ball: 6.289587
Racket: 45.83272
Blocks: 8.100111
Slow: 1.150546
Fast: 2.966203
Enlarge: 0.3838165
Shrink: 0
Destructor: 0
Sticky: 0.8160867
Level 1 Gaze Logging
Ball: 11.60126
Racket: 40.49958
Blocks: 80.00227
Slow: 3.28428
Fast: 0.2815178
Enlarge: 2.081458
Shrink: 1.666226
Destructor: 1.100046
Sticky: 0
Participant 1 – Medium:
Level 1 Gaze Logging
Ball: 18.94375
Racket: 42.89381
Blocks: 5.799256
Slow: 3.769356
Fast: 0.1673606
Enlarge: 0.0820847
Shrink: 0.6163186
Destructor: 1.000601
Sticky: 2.317075
Level 1 Gaze Logging
Ball: 13.51487
Racket: 43.94612
Blocks: 80.0102
Slow: 0
Fast: 1.0674
Enlarge: 0.3834044
Shrink: 0.8500188
Destructor: 2.250339
Sticky: 0.2326647
Level 1 Gaze Logging
Ball: 10.38087
Racket: 41.88626
Blocks: 15.5827
Slow: 0.7815382
Fast: 2.650429
Enlarge: 1.582641
Shrink: 0.3993519
Destructor: 0
Sticky: 0.0165161
Partcipant 1 – fast:
Level 1 Gaze Logging
Ball: 31.49242
Racket: 8.40009
Blocks: 59.2073
Slow: 5.470974
Fast: 1.300836
Enlarge: 0.4004512
Shrink: 3.098726
Destructor: 0.7323367
Sticky: 0.4500434
Level 1 Gaze Logging
Ball: 32.94662
Racket: 73.21638
Blocks: 22.3974
Slow: 5.647902
Fast: 2.284136
Enlarge: 2.383301
Shrink: 0
Destructor: 1.899534
Sticky: 2.200315
Level 1 Gaze Logging
Ball: 32.84812
Racket: 92.92934
Blocks: 13.24673
Slow: 3.115646
Fast: 1.799756
Enlarge: 4.084972
Shrink: 1.434316
Destructor: 0
Sticky: 0.2168911
Partcipant 2 – slow:
Level 1 Gaze Logging
Ball: 5.16715
Racket: 0
Blocks: 19.997
Slow: 0.4325026
Fast: 1.517021
Enlarge: 0.4988995
Shrink: 1.48338
Destructor: 0
Sticky: 0
Level 1 Gaze Logging
Ball: 8.407804
Racket: 0
Blocks: 80.01009
Slow: 0.3164963
Fast: 0
Enlarge: 0.61663
Shrink: 0
Destructor: 0
Sticky: 0
Level 1 Gaze Logging
Ball: 4.493901
Racket: 0
Blocks: 15.90282
Slow: 0.2501278
Fast: 0.7001745
Enlarge: 0
Shrink: 0.250158
Destructor: 0.0168506
Sticky: 0
Participant 2 – medium:
Level 1 Gaze Logging
Ball: 9.004568
Racket: 0
Blocks: 19.34712
Slow: 0.9669013
Fast: 1.449938
Enlarge: 0.3662705
Shrink: 0.9336085
Destructor: 0.6002329
Sticky: 0
Level 1 Gaze Logging
Ball: 9.738883
Racket: 0
Blocks: 15.33011
Slow: 0.4505052
Fast: 0.4002005
Enlarge: 0.6008443
Shrink: 1.366842
Destructor: 0.7009124
Sticky: 0
Level 1 Gaze Logging
Ball: 4.41079
Racket: 0
Blocks: 80.01275
Slow: 1.015855
Fast: 0.4995959
Enlarge: 0.4338197
Shrink: 1.466489
Destructor: 0.4500207
Sticky: 1.098477
Participant 2 – fast:
Level 1 Gaze Logging
Ball: 15.83397
Racket: 7.986393
Blocks: 79.71999
Slow: 0.8169758
Fast: 0.6997507
Enlarge: 1.134616
Shrink: 1.449276
Destructor: 0.3340891
Sticky: 0.1833477
Level 1 Gaze Logging
Ball: 18.57607
Racket: 0
Blocks: 61.13648
Slow: 1.698052
Fast: 0.7329782
Enlarge: 0.500637
Shrink: 2.599958
Destructor: 1.316194
Sticky: 0
Level 1 Gaze Logging
Ball: 19.33537
Racket: 10.08981
Blocks: 17.01433
Slow: 4.215042
Fast: 0.4163074
Enlarge: 0.099514
Shrink: 0.1662122
Destructor: 0.1004815
Sticky: 0
Participant 3 – slow:
Level 1 Gaze Logging
Ball: 13.39751
Racket: 13.36672
Blocks: 79.90398
Slow: 0.8338404
Fast: 1.299283
Enlarge: 1.166661
Shrink: 0
Destructor: 0
Sticky: 0
Level 1 Gaze Logging
Ball: 22.5318
Racket: 36.81059
Blocks: 62.93246
Slow: 0.1833613
Fast: 0.5335769
Enlarge: 0
Shrink: 1.099615
Destructor: 0.3662709
Sticky: 0
Level 1 Gaze Logging
Ball: 15.60279
Racket: 21.49318
Blocks: 18.39465
Slow: 1.55021
Fast: 0
Enlarge: 0
Shrink: 0.0326381
Destructor: 1.882113
Sticky: 0
Participant 3 – medium:
Level 1 Gaze Logging
Ball: 14.86653
Racket: 27.7726
Blocks: 79.90369
Slow: 2.032989
Fast: 1.183092
Enlarge: 3.464406
Shrink: 1.315056
Destructor: 0
Sticky: 0
Level 1 Gaze Logging
Ball: 19.48988
Racket: 41.68676
Blocks: 6.136158
Slow: 3.015706
Fast: 1.216265
Enlarge: 1.702738
Shrink: 2.016008
Destructor: 0
Sticky: 0.9842538
Level 1 Gaze Logging
Ball: 17.25334
Racket: 75.47806
Blocks: 58.56461
Slow: 2.118085
Fast: 0.6342602
Enlarge: 0.5832416
Shrink: 2.715107
Destructor: 0.3002183
Sticky: 0.316857
Participant 3 – fast:
Level 1 Gaze Logging
Ball: 47.13136
Racket: 46.78366
Blocks: 8.081146
Slow: 1.250741
Fast: 0.9828895
Enlarge: 0.8182472
Shrink: 1.948415
Destructor: 0.9507852
Sticky: 0
Level 1 Gaze Logging
Ball: 49.06231
Racket: 60.46688
Blocks: 80.00274
Slow: 1.398796
Fast: 0.9342638
Enlarge: 2.167299
Shrink: 1.033012
Destructor: 0.3497688
Sticky: 0.6339062
Level 1 Gaze Logging
Ball: 24.80988
Racket: 133.86
Blocks: 67.83218
Slow: 1.264122
Fast: 0.682645
Enlarge: 1.800957
Shrink: 0.2004393
Destructor: 5.700058
Sticky: 0.4338903
Participant 4 – slow:
Level 1 Gaze Logging
Ball: 20.09122
Racket: 66.06065
Blocks: 7.332118
Slow: 0.7836357
Fast: 0.5831311
Enlarge: 1.382845
Shrink: 0.1834548
Destructor: 0
Sticky: 0
Level 1 Gaze Logging
Ball: 14.47509
Racket: 37.06368
Blocks: 6.984249
Slow: 1.684294
Fast: 0.335488
Enlarge: 0.3993584
Shrink: 0.4669737
Destructor: 0
Sticky: 0
Level 1 Gaze Logging
Ball: 17.1918
Racket: 46.9268
Blocks: 80.00251
Slow: 0.2343945
Fast: 0
Enlarge: 0
Shrink: 2.949878
Destructor: 0
Sticky: 0
Participant 4 – medium:
Level 1 Gaze Logging
Ball: 6.033322
Racket: 68.67223
Blocks: 5.449038
Slow: 0.3173081
Fast: 0.2159255
Enlarge: 0
Shrink: 0.2670631
Destructor: 0
Sticky: 0
Level 1 Gaze Logging
Ball: 33.31586
Racket: 93.9619
Blocks: 1.749849
Slow: 0.7990263
Fast: 1.132668
Enlarge: 0.0840389
Shrink: 0.1357304
Destructor: 0.8833668
Sticky: 0.6990635
Level 1 Gaze Logging
Ball: 25.46096
Racket: 66.86402
Blocks: 80.00698
Slow: 0
Fast: 0.333962
Enlarge: 0.799778
Shrink: 2.532837
Destructor: 0
Sticky: 0
Participant 4 – fast:
Level 1 Gaze Logging
Ball: 38.70448
Racket: 120.285
Blocks: 4.865451
Slow: 3.150663
Fast: 0.2339665
Enlarge: 1.549518
Shrink: 0.3491181
Destructor: 0.0333123
Sticky: 0
Level 1 Gaze Logging
Ball: 45.92964
Racket: 116.6909
Blocks: 80.00664
Slow: 1.216228
Fast: 1.850111
Enlarge: 0.4171427
Shrink: 0.1333098
Destructor: 0.5502037
Sticky: 0.0337018
Level 1 Gaze Logging
Ball: 30.90442
Racket: 79.86823
Blocks: 48.50605
Slow: 0.1993779
Fast: 1.91812
Enlarge: 0.4316317
Shrink: 1.001471
Destructor: 0.5669097
Sticky: 0.5004727
Participant 5 – slow:
Level 1 Gaze Logging
Ball: 10.92375
Racket: 75.93484
Blocks: 5.924689
Slow: 0.6990487
Fast: 0
Enlarge: 0.1677857
Shrink: 0.2840989
Destructor: 0
Sticky: 0
Level 1 Gaze Logging
Ball: 9.893981
Racket: 22.17497
Blocks: 80.00244
Slow: 2.887215
Fast: 0.7486459
Enlarge: 0
Shrink: 0.0339605
Destructor: 0.666799
Sticky: 0
Level 1 Gaze Logging
Ball: 8.748375
Racket: 24.37075
Blocks: 4.780408
Slow: 2.300848
Fast: 3.033147
Enlarge: 1.567938
Shrink: 0.8661237
Destructor: 0
Sticky: 0
Participant 5 – medium:
Level 1 Gaze Logging
Ball: 16.27177
Racket: 56.53201
Blocks: 53.45752
Slow: 0.9998881
Fast: 1.86753
Enlarge: 2.084124
Shrink: 2.465982
Destructor: 1.066103
Sticky: 0.450186
Level 1 Gaze Logging
Ball: 20.94051
Racket: 46.65269
Blocks: 80.0137
Slow: 2.596123
Fast: 1.132361
Enlarge: 0.1666943
Shrink: 0
Destructor: 1.065641
Sticky: 0
Level 1 Gaze Logging
Ball: 17.71045
Racket: 33.02407
Blocks: 63.42382
Slow: 3.201142
Fast: 2.983593
Enlarge: 0
Shrink: 1.065506
Destructor: 0.1832867
Sticky: 0
Participant 5 – fast:
Level 1 Gaze Logging
Ball: 36.49709
Racket: 41.98351
Blocks: 17.53145
Slow: 3.236687
Fast: 0.6998728
Enlarge: 3.164567
Shrink: 1.783449
Destructor: 0.2827328
Sticky: 1.484315
Level 1 Gaze Logging
Ball: 19.92422
Racket: 31.21228
Blocks: 11.3348
Slow: 2.88246
Fast: 1.031228
Enlarge: 0.4498506
Shrink: 1.033586
Destructor: 0.1661672
Sticky: 0
Level 1 Gaze Logging
Ball: 29.74369
Racket: 80.89175
Blocks: 80.01077
Slow: 2.571244
Fast: 4.184511
Enlarge: 3.563781
Shrink: 1.000167
Destructor: 1.032236
Sticky: 0.2480672
Participant 6 – slow:
Level 1 Gaze Logging
Ball: 5.762348
Racket: 25.52998
Blocks: 79.90336
Slow: 0.0667832
Fast: 0.8166052
Enlarge: 0
Shrink: 0.1002014
Destructor: 0.08323471
Sticky: 0
Level 1 Gaze Logging
Ball: 8.010356
Racket: 26.19433
Blocks: 5.198756
Slow: 0.282753
Fast: 0
Enlarge: 0.6835723
Shrink: 0.1671642
Destructor: 0
Sticky: 0
Level 1 Gaze Logging
Ball: 2.449667
Racket: 25.90156
Blocks: 56.69009
Slow: 0.6830505
Fast: 0.8823397
Enlarge: 0
Shrink: 0.0333528
Destructor: 0.3165933
Sticky: 0
Participant 6 – medium:
Level 1 Gaze Logging
Ball: 4.882896
Racket: 31.05025
Blocks: 3.432292
Slow: 0.899954
Fast: 0.0667287
Enlarge: 0
Shrink: 0.1000329
Destructor: 0
Sticky: 0
Level 1 Gaze Logging
Ball: 8.217857
Racket: 26.29244
Blocks: 80.00655
Slow: 0.0333997
Fast: 0.9134892
Enlarge: 0.5832729
Shrink: 0.0332779
Destructor: 0.033465
Sticky: 0
Level 1 Gaze Logging
Ball: 7.870317
Racket: 57.54198
Blocks: 7.29739
Slow: 2.23294
Fast: 0.1666537
Enlarge: 0.5833737
Shrink: 0
Destructor: 0.0503647
Sticky: 0
Participant 6 – fast:
Level 1 Gaze Logging
Ball: 8.638738
Racket: 69.07999
Blocks: 61.35606
Slow: 1.549967
Fast: 0.7005693
Enlarge: 0
Shrink: 0.5658283
Destructor: 0
Sticky: 0
Level 1 Gaze Logging
Ball: 12.74884
Racket: 61.85991
Blocks: 80.0077
Slow: 1.151771
Fast: 0.1673731
Enlarge: 0.2499718
Shrink: 1.415162
Destructor: 0.1169666
Sticky: 0
Level 1 Gaze Logging
Ball: 15.03252
Racket: 120.474
Blocks: 37.84387
Slow: 0.06644219
Fast: 0.3333974
Enlarge: 0.1157416
Shrink: 2.166272
Destructor: 0.0832302
Sticky: 0
Participant 7 – slow:
Level 1 Gaze Logging
Ball: 11.4158
Racket: 17.56759
Blocks: 10.99857
Slow: 0.5168582
Fast: 0.4487092
Enlarge: 1.150221
Shrink: 2.36504
Destructor: 0.4172213
Sticky: 0
Level 1 Gaze Logging
Ball: 15.81886
Racket: 24.63872
Blocks: 20.24716
Slow: 0.7660826
Fast: 1.581759
Enlarge: 0.2002085
Shrink: 0.0999067
Destructor: 0.5665728
Sticky: 0
Level 1 Gaze Logging
Ball: 11.65063
Racket: 27.76486
Blocks: 80.00526
Slow: 3.218412
Fast: 0.1997543
Enlarge: 0.8331487
Shrink: 0.1504697
Destructor: 0.0663428
Sticky: 0.8998984
Participant 7 – medium:
Level 1 Gaze Logging
Ball: 7.659086
Racket: 22.40442
Blocks: 35.11121
Slow: 0
Fast: 0.9845469
Enlarge: 0.0326443
Shrink: 0.6323675
Destructor: 0.2004961
Sticky: 0
Level 1 Gaze Logging
Ball: 12.16181
Racket: 24.81977
Blocks: 29.92715
Slow: 1.466349
Fast: 1.599524
Enlarge: 0.7829133
Shrink: 0.3664988
Destructor: 0.4997461
Sticky: 0
Level 1 Gaze Logging
Ball: 8.188521
Racket: 38.10017
Blocks: 80.01109
Slow: 0.2171079
Fast: 0.583716
Enlarge: 0.1834325
Shrink: 0.0832498
Destructor: 0.4500753
Sticky: 0
Partcipant 7 – fast:
Level 1 Gaze Logging
Ball: 20.90649
Racket: 102.2894
Blocks: 3.149002
Slow: 1.664937
Fast: 1.216445
Enlarge: 1.332771
Shrink: 0.216484
Destructor: 0.0992367
Sticky: 0
Level 1 Gaze Logging
Ball: 12.99492
Racket: 76.39504
Blocks: 80.0144
Slow: 1.18353
Fast: 2.216555
Enlarge: 0.2328176
Shrink: 1.801541
Destructor: 0.116574
Sticky: 0
Level 1 Gaze Logging
Ball: 11.42855
Racket: 32.60067
Blocks: 32.79946
Slow: 2.350371
Fast: 0.1991624
Enlarge: 0.1001905
Shrink: 0.8668844
Destructor: 0.1662654
Sticky: 0
Participant 8 – slow:
Level 1 Gaze Logging
Ball: 1.331248
Racket: 32.60218
Blocks: 3.551302
Slow: 0
Fast: 0.7499442
Enlarge: 0.1165165
Shrink: 0
Destructor: 0
Sticky: 0
Level 1 Gaze Logging
Ball: 2.742418
Racket: 47.84989
Blocks: 80.006
Slow: 0.4167093
Fast: 0
Enlarge: 0
Shrink: 0
Destructor: 0.2004816
Sticky: 0
Level 1 Gaze Logging
Ball: 3.413892
Racket: 32.43032
Blocks: 47.85797
Slow: 0
Fast: 0.0166463
Enlarge: 0.5830404
Shrink: 0
Destructor: 0.399938
Sticky: 0
Participant 8 – medium:
Level 1 Gaze Logging
Ball: 12.98289
Racket: 36.27109
Blocks: 79.95308
Slow: 0.1672215
Fast: 0.5161715
Enlarge: 1.148201
Shrink: 0.250504
Destructor: 0.0336961
Sticky: 0
Level 1 Gaze Logging
Ball: 5.829711
Racket: 38.96655
Blocks: 21.54684
Slow: 0
Fast: 0
Enlarge: 1.333836
Shrink: 0.0664073
Destructor: 0.1163423
Sticky: 0
Level 1 Gaze Logging
Ball: 3.399004
Racket: 48.82169
Blocks: 1.131288
Slow: 0.4818965
Fast: 0.419059
Enlarge: 0
Shrink: 0
Destructor: 0
Sticky: 0
Participant 8 – fast:
Level 1 Gaze Logging
Ball: 18.95342
Racket: 113.4843
Blocks: 79.90336
Slow: 0.5676556
Fast: 0.5991066
Enlarge: 0.4667195
Shrink: 0.065514
Destructor: 0
Sticky: 0.0337237
Level 1 Gaze Logging
Ball: 10.3534
Racket: 97.62351
Blocks: 2.232723
Slow: 0.9334871
Fast: 1.797069
Enlarge: 0.1829404
Shrink: 0.265853
Destructor: 0.4337209
Sticky: 0
Level 1 Gaze Logging
Ball: 15.13697
Racket: 94.81552
Blocks: 56.37408
Slow: 0.4814266
Fast: 0.0335142
Enlarge: 0.5011484
Shrink: 0.3328279
Destructor: 0.4504337
Sticky: 0
Participant 9 – slow:
Level 1 Gaze Logging
Ball: 11.44697
Racket: 23.79382
Blocks: 5.999184
Slow: 1.149784
Fast: 0.0334366
Enlarge: 0.9334643
Shrink: 1.165761
Destructor: 0
Sticky: 0
Level 1 Gaze Logging
Ball: 7.308985
Racket: 18.69383
Blocks: 42.20686
Slow: 1.466367
Fast: 0
Enlarge: 0.1329726
Shrink: 1.84825
Destructor: 0.2498755
Sticky: 0
Level 1 Gaze Logging
Ball: 5.678749
Racket: 21.99479
Blocks: 80.00219
Slow: 1.74885
Fast: 1.533193
Enlarge: 0.8826805
Shrink: 0
Destructor: 0
Sticky: 0.1337693
Participant 9 – medium:
Level 1 Gaze Logging
Ball: 9.952309
Racket: 25.70806
Blocks: 79.88667
Slow: 1.918098
Fast: 3.798785
Enlarge: 0.1163619
Shrink: 0.2337773
Destructor: 0
Sticky: 0
Level 1 Gaze Logging
Ball: 7.305156
Racket: 27.39979
Blocks: 27.76144
Slow: 1.481549
Fast: 0.7992561
Enlarge: 0.6336225
Shrink: 0.7829956
Destructor: 0.033341
Sticky: 0.1504655
Level 1 Gaze Logging
Ball: 4.182069
Racket: 29.92155
Blocks: 7.532586
Slow: 0.4498656
Fast: 0.9164044
Enlarge: 1.383181
Shrink: 1.517018
Destructor: 0
Sticky: 0.2496184
Participant 9 – fast:
Level 1 Gaze Logging
Ball: 19.29976
Racket: 83.0322
Blocks: 59.33986
Slow: 2.816068
Fast: 1.350517
Enlarge: 0.2324947
Shrink: 0.0331377
Destructor: 0.4676907
Sticky: 0.5330722
Level 1 Gaze Logging
Ball: 18.10292
Racket: 50.2337
Blocks: 80.00499
Slow: 0.4176506
Fast: 1.616292
Enlarge: 0.5495446
Shrink: 2.998482
Destructor: 0.0504447
Sticky: 0.71663
Level 1 Gaze Logging
Ball: 28.18307
Racket: 63.61323
Blocks: 5.067404
Slow: 3.485107
Fast: 0.6669638
Enlarge: 1.631471
Shrink: 2.898001
Destructor: 0.1497366
Sticky: 0
Participant 10 – slow:
Level 1 Gaze Logging
Ball: 19.64212
Racket: 20.28986
Blocks: 79.95374
Slow: 1.632546
Fast: 0
Enlarge: 2.955282
Shrink: 0.7669034
Destructor: 1.931401
Sticky: 0
Level 1 Gaze Logging
Ball: 21.44183
Racket: 26.42944
Blocks: 14.25012
Slow: 0.2658803
Fast: 2.615088
Enlarge: 0.8498118
Shrink: 1.216327
Destructor: 0.1836949
Sticky: 1.068611
Level 1 Gaze Logging
Ball: 36.04441
Racket: 21.6061
Blocks: 11.06443
Slow: 3.881307
Fast: 1.032955
Enlarge: 0
Shrink: 0.8499538
Destructor: 0
Sticky: 0
Participant 10 – medium:
Level 1 Gaze Logging
Ball: 12.10839
Racket: 82.86877
Blocks: 79.90364
Slow: 3.616464
Fast: 0.6006816
Enlarge: 0.1170498
Shrink: 0.816817
Destructor: 0
Sticky: 0
Level 1 Gaze Logging
Ball: 16.90644
Racket: 52.30312
Blocks: 41.21085
Slow: 1.900996
Fast: 1.133301
Enlarge: 0
Shrink: 4.798335
Destructor: 0
Sticky: 0
Level 1 Gaze Logging
Ball: 24.90568
Racket: 50.64729
Blocks: 8.047902
Slow: 1.4011
Fast: 0.6495919
Enlarge: 1.632385
Shrink: 0.4835296
Destructor: 1.000572
Sticky: 0
Participant 10 – fast:
Level 1 Gaze Logging
Ball: 54.94739
Racket: 62.66701
Blocks: 57.34034
Slow: 1.717786
Fast: 1.699365
Enlarge: 2.11679
Shrink: 0
Destructor: 0.9829948
Sticky: 0.7658422
Level 1 Gaze Logging
Ball: 45.62025
Racket: 67.95138
Blocks: 80.00251
Slow: 1.932049
Fast: 1.233037
Enlarge: 0.6832553
Shrink: 0
Destructor: 0
Sticky: 0.416629
Level 1 Gaze Logging
Ball: 55.56848
Racket: 50.47579
Blocks: 57.25661
Slow: 2.648495
Fast: 0.7495916
Enlarge: 0.5833156
Shrink: 2.599191
Destructor: 0.6502873
Sticky: 0
Participant 11 – slow:
Level 1 Gaze Logging
Ball: 22.62223
Racket: 26.48476
Blocks: 22.53077
Slow: 5.066495
Fast: 0
Enlarge: 0.0333037
Shrink: 0.7985107
Destructor: 1.499757
Sticky: 1.415226
Level 1 Gaze Logging
Ball: 27.28844
Racket: 40.16528
Blocks: 80.00893
Slow: 2.682215
Fast: 1.666193
Enlarge: 5.503611
Shrink: 1.65126
Destructor: 0.1506102
Sticky: 0
Level 1 Gaze Logging
Ball: 18.09153
Racket: 11.45298
Blocks: 55.47356
Slow: 0.283357
Fast: 0.5338491
Enlarge: 0.8676678
Shrink: 2.098196
Destructor: 0.9497247
Sticky: 0.6343929
Participant 11 – medium:
Level 1 Gaze Logging
Ball: 22.62223
Racket: 26.48476
Blocks: 22.53077
Slow: 5.066495
Fast: 0
Enlarge: 0.0333037
Shrink: 0.7985107
Destructor: 1.499757
Sticky: 1.415226
Level 1 Gaze Logging
Ball: 27.28844
Racket: 40.16528
Blocks: 80.00893
Slow: 2.682215
Fast: 1.666193
Enlarge: 5.503611
Shrink: 1.65126
Destructor: 0.1506102
Sticky: 0
Level 1 Gaze Logging
Ball: 18.09153
Racket: 11.45298
Blocks: 55.47356
Slow: 0.283357
Fast: 0.5338491
Enlarge: 0.8676678
Shrink: 2.098196
Destructor: 0.9497247
Sticky: 0.6343929
Partcipant 11 – fast:
Level 1 Gaze Logging
Ball: 41.36404
Racket: 97.41636
Blocks: 25.23033
Slow: 5.281194
Fast: 3.381054
Enlarge: 0.5004089
Shrink: 1.183462
Destructor: 2.166671
Sticky: 0
Level 1 Gaze Logging
Ball: 45.32401
Racket: 75.98265
Blocks: 30.97778
Slow: 3.231428
Fast: 3.733148
Enlarge: 3.96674
Shrink: 2.065669
Destructor: 0.2668878
Sticky: 1.232843
Level 1 Gaze Logging
Ball: 46.86933
Racket: 47.54942
Blocks: 80.0077
Slow: 3.266431
Fast: 0.433123
Enlarge: 4.547896
Shrink: 0.1003677
Destructor: 0.1334698
Sticky: 0.7656317
Participant 12 – slow:
Level 1 Gaze Logging
Ball: 11.4115
Racket: 5.063105
Blocks: 13.52827
Slow: 2.445259
Fast: 0.9318228
Enlarge: 0.8835701
Shrink: 2.065989
Destructor: 1.066896
Sticky: 0
Level 1 Gaze Logging
Ball: 8.14743
Racket: 12.38761
Blocks: 80.00331
Slow: 0.2166822
Fast: 0.7167906
Enlarge: 0.0327854
Shrink: 0.5325366
Destructor: 0
Sticky: 0
Level 1 Gaze Logging
Ball: 14.24117
Racket: 13.00076
Blocks: 18.76003
Slow: 1.031622
Fast: 0
Enlarge: 1.265518
Shrink: 0.0662475
Destructor: 0
Sticky: 0.399987
Participant 12 – medium:
Level 1 Gaze Logging
Ball: 24.90568
Racket: 50.64729
Blocks: 8.047902
Slow: 1.4011
Fast: 0.6495919
Enlarge: 1.632385
Shrink: 0.4835296
Destructor: 1.000572
Sticky: 0
Level 1 Gaze Logging
Ball: 27.26564
Racket: 21.82124
Blocks: 79.8867
Slow: 6.002906
Fast: 0.6172082
Enlarge: 1.997235
Shrink: 0.3498529
Destructor: 0
Sticky: 0.1505951
Level 1 Gaze Logging
Ball: 33.09294
Racket: 49.56408
Blocks: 13.96683
Slow: 1.699397
Fast: 0
Enlarge: 0.4166989
Shrink: 3.254223
Destructor: 0.1821044
Sticky: 0.3332804
Participant 12 – fast:
Level 1 Gaze Logging
Ball: 51.03405
Racket: 75.16602
Blocks: 79.90376
Slow: 4.64929
Fast: 0
Enlarge: 1.066603
Shrink: 3.034614
Destructor: 0.1832571
Sticky: 0
Level 1 Gaze Logging
Ball: 55.40553
Racket: 46.2388
Blocks: 10.76914
Slow: 2.94781
Fast: 0.6997057
Enlarge: 4.166424
Shrink: 0.9663248
Destructor: 0.6997743
Sticky: 0
Level 1 Gaze Logging
Ball: 36.40891
Racket: 46.71301
Blocks: 42.02892
Slow: 3.149061
Fast: 2.09996
Enlarge: 2.899125
Shrink: 1.765733
Destructor: 2.099781
Sticky: 0.6156248
Evaluation data analysis:
Disclaimer: All numbers here are the average of the raw data in this link:https://drive.google.com/drive/folders/1mYRvpkvEPytdrRP6BhV4CSlZxqPxjKIA
Slow Average:5 second spawn rate on powerups:
Par. ball racket block slow fast enlarge shrink destructor sticky Score
1 8.75s 37,2s 33,81s 1,70 1,41s 1,35s 0,55s 0,36s 0,27s 3200
2 6,01s 0s 38,63s 0.33s 0,73s 0,36s 0,55s 0,003s 0s 7900
3 17,17s 23,88s 53,74s 0,85s 0,60s 1,16s 0,37s 0,74s 0s -3500
4 17,25s 50,01 31,33s 0,89s 0,30s 0,59s 0,52s 0s 0s 1700
5 9,85s 40,82s 30,23s 1,95s 1,25s 0,57s 0,39s 0,22s 0s 5000
6 5,4s 25,87s 47,26s 0,34s 0,56s 0,22s 0,09s 0,13s 0s 4300
7 12,95 23,31s 37,07s 1,49s 0,73s 0,72s 0,86s 0,34s 0,29s 5900
8 2,49 37,62 43,8 0,13 0,25 0,23 0s 0,19 0 -800
9 8.13 21,49 42,73 1,44 0,52 0,64 1 0,08 0,04 -2400
10 25,70 22,76 35,08 1,92 1,21 1,26 0,93 0,70 0,35 4500
11 22,66 26,03 52,66 2,67 0,73 2,13 1,51 0,86 0,68 4800
12 28,41 40,67 33,96 3,03 0,41 1,34 1,35 0,39 0,16 6100
Medium Average:3.5 second spawnrate on powerups:
Par. ball racket block slow fast enlarge shrink destructor sticky Score
1 14,27s 42,90s 33,79s 1,51s 1,29s 0,68s 0,61s 1,08 0.85s 4200
2 7,71s 0s 38,22s 0,79s 0,77s 0,46s 1,25s 1,75s 0,36s 6900
3 17,19s 48,30s 48,19s 2,38s 1,00s 1,91s 2,01s 0,1s 0,43s -4500
4 21,6s 76,49s 29,06s 0,36s 0,54s 0,29s 0,97s 0,29s 0,23s 5700
5 18,30s 45,4s 65,62s 2,23s 1,99s 0,74s 1,17s 0,76 0,15s 4300
6 6,98s 8,70s 30,24s 1,05s 0,37s 0,38s 0,04s 0,02s 0s 1700
7 9,33s 28,43s 48,34s 0,55s 1,05s 0,33 0,35s 0,38 0s 5700
8 7,39 41,35 34,20 0,21 0,30 0,82 0,31 0,04 0 1700
9 7,14 27,67 38,39 1,27 1,83 0,75 0,84 0,01 0,13 -12100
10 17,96 61,93 43,05 2,30 0,79 0,58 2,02 0,33 0 2400
11 22,66 26,03 52,66 2,67 0,73 2,13 0,84 0,86 0,68 1600
12 28,41 40,67 33,96 3,09 0,41 1,34 1,35 0,39 0,16 11000
Fast Average:2 second spawn rate on powerups:
Par. ball racket block slow fast enlarge shrink destructor sticky Score
1 32,42s 58,17s 31,61s 14.1s 1.76s 1,36s 1,49s 0,83s 0,93s 6900
2 17,91s 6,02s 52,61s 2,23s 14,14s 0,57s 1,39s 0.58s 0.06s 3400
3 40,33s 80,36s 51,97s 1,3s 2,59s 1,59s 1,05s 2,33s 1,68s 1300
4 38,50s 105,61s 44,45s 1,51s 1,33s 0,79s 0,49s 0,38s 0,17s 8800
5 28,71s 51,36s 36,29s 2,89s 1,96s 2,38s 1,27s 0,49s 0,57s 4700
6 12,13s 83,79s 59,73s 0,91s 0,39s 0,11s 1,37s 0,06s 0s -3000
7 15,10s 70,42s 38,64s 1,73s 1,20s 0,16s 0,95s 0,12 0s 7600
8 14,81s 101,97 46,16s 0,65 0,80 0,38 0,21 0,29 0,01 1200
9 21,85 65,62 48,13 2,23 3,57 0,8 1,97 0,21 0,40 1200
10 52,04 60,36 64,86 2,09 1,22 1,12 0,86 0,54 0,39 5500
11 44,51 73,64 45,4 3,91 2,51 3 1,11 0,85 0,66 2000
12 30,94 56,03 44,22 3,57 0,92 2,70 1,91 0,98 0,20 7500
Ultimate averages:
Averages of all participants performance in each segment:
Slow SP ball paddle block slow fast enlarge
shrink destructor sticky score
Count,N:
12 12 12 12 12 12 12 12 12 12
Sum, Σx 164.77 349.66 480.3 16.74 8.7 10.57 8.12 4.013 1.79 36700
Mean, μ: 13,73s 29,13s 40,025s
1,39s 0,72s 0,88s 0,67s 0,33s 0,14s 3058,33
Std.Deviation, σ:
8.113 12.490 7.658 0.884 0.361 0.548 0.445 0.279 0.204 3424.776
Variance, σ2:
65.825 156.0151
58.645 0.782 0.130 0.300 0.198 0.078 0.041 11729097.222
MediumSP
ball paddle block slow fast enlarge
shrink destructor sticky score
Count,N:
12 12 12 12 12 12 12 12 12 12
Sum, Σx 178.94 536.97 495.72 39.2 11.07 10.41 12.75 6.01 2.99 28600
Mean, μ: 14,91s 37,32s 41,31s 3.015s
0.922 0.867 0,98s 0,5s 0,249s 2383,33
Std.Deviation, σ:
6.929 23.947 10.293 5.272 0.524 0.583 0.573 0.503 0.269 5644.441
Variance, σ2: 48.021
573.489 105.949
27.804
0.275 0.340 0.328 0.253 0.072 31859722.22
Fast SP ball paddle block slow fast enlarge
shrink destructor sticky score
Count, N: 12 12 12 12 12 12 12 12 12 12
Sum, Σx349.25 813.35
564.07 37.12 32.39 14.96 14.07 7.66 5.07 47100
Mean, μ: 29,10s 67,77s 47,00s 3,09s 2,69s 1,24s 1,17s 0,63s 0,42s 3925
Std.Deviation, σ:
12.445 24.883 9.025 3.450 3.553 0.944 0.494 0.582 0.473 3362.321
Variance, σ2:
154.892
619.166 81.452 11.908
12.628
0.892 0.244 0.339 0.224 11305208.333
● They looked more at every single object and yielded the highest score at the fastestspawn-rate
● They generally looked more at the objects around them in the medium spawn rate,but got a lower average score than the low spawn-rate
○ There was an outlier that managed to get -12100○ Participant 3 was “trolling” in their first run, which was “medium, slow, fast”
Participant scores:
Participant 1 - 3200 slow - 4200 medium - 6900 fast 14300 - Did not have knowledge of power-ups or core game mechanics (Gradual speed increase, score death decrease, timed level) during slow segment Participant 2 - 7900 Slow - 6900 Medium - 3400 fast 17800 Participant 3 - -3500 Slow - -4500 Medium - 1300 Fast - messed around during 1st level of 1st run -5700 Participant 4 - 1700 Slow - 5700 Medium - 8800 Fast 16200 Participant 5 - 5000 Slow - 4300 Medium - 4700 Fast Participant 6 - -3000 Fast - 1700 Medium - 4300 Slow 3000 Participant 7 - 5900 Slow - 5700 Medium - 7600 Fast 19200 Participant 8 - -800 Slow - 1700 Fast
- 1200 Medium 2300 Participant 9 - -12100 Medium - -2400 Slow - 1200 Fast -13300 Participant 10 - 4500 Slow - 2400 Medium - 5500 Fast 12400 Participant 11 - 2000 Fast - 4800 Slow - 1600 Medium 8400 Participant 12 - 7500 Fast - 11000 Medium - 6100 Slow 24700
B Appendix B
B.1 Two-way ANOVA data
44
Two-way ANOVA data
Data calculated from the two-way ANOVA test:
Scores and gaze logging:
Measure Variance (Slow) Variance (Medium) Variance (Fast) P-value
Scores 1279 3475 1233 0.4678829001
Ball 71.80911742 52.38688788 168.9734629 0.0000005
Racket 170.1983788 445.0811841 675.4546629 0.0000021
Slow Ball 5.333011364 1.93990569 12.99073333 0.3384662064
Fast Ball 0.8923295455 0.6122495362 13.77651742 0.3156764545
Small Racket 1.355151515 0.7966604824 0.2668022727 0.2842405788
Large Racket 2.050961174 0.7569990724 0.9734060606 0.001138442212
Destructor 4.083688447 0.5635791589 0.3699787879 0.1504905714
Sticky 0.2852793561 0.1618351886 0.2445840909 0.9095531916
Brick 63.97671818 115.5807636 88.85675379 0.1703812528
iGEQ questionnaire:
Measure Variance (Slow) Variance (Medium) Variance (Fast) P-value
Success Feeling 2.27 1.45 2.02 0.90
Bored 0.79 0.97 0.42 0.13
Impressive 1.30 1.36 0.99 0.66
Forgot Surroundings 1.17 1.90 1.90 0.86
Frustration 1.45 1.24 1.24 0.86
Tiresome 0.73 0.88 1.17 0.52
Irritable 1.90 1.70 2.15 0.34
Skillful 1.52 1.36 1.24 0.81
Absorbed 2.73 1.84 2.09 0.29
Contentment 0.81 0.79 1.48 0.81
Challenge 0.55 0.27 0.45 0.90
Stimulated 1.33 1.00 0.97 0.75
Felt Good 1.88 1.90 1.33 0.94
NASA task load manager:
Mental Demand 1.424242424 0.9318181818 1.151515152 0.1573148387
Physical Demand 0.9090909091 1.174242424 1.356060606 0.1765911004
Task Pace 0.6287878788 0.9924242424 0.7272727273 0.8312054341
Feeling of Success 1.787878788 1.295454545 0.9924242424 0.9120588647
Workload 0.2424242424 0.2727272727 0.5151515152 0.2737299079
Negative Emotion 1.901515152 1.696969697 2.151515152 0.3386321712
C Appendix C
C.1 Code snippets
47
Code snippets:
Paddle movement:
Ball movement: